This paper investigates the importance of including paranasal sinuses in the computational modeling of the nasal airflow. Three dimensional models of human nasal airway with and without including paranasal sinuses were reconstructed from Computed Tomography (CT) axial images of a subject with healthy nasal airway. The reconstruction process was performed using MIMICS® software program. The airway volume was discretized using TGRID® mesh generator. Steady Reynolds-Averaged Navier-Stokes (RANS) simulations were carried in both inspiratory and expiratory phases of respiratory cycle at a peak flow rate of 15 L/min in FLUENT®. The results show that the left and right nasal resistances change with less than 11% when paranasal sinuses are included in the computational model of the nasal airway. The flow into the sinuses is characterized by very low velocities during both inspiration and expiration conditions. The velocity distributions in the main nasal passage show small change predominantly in regions closer to the paranasal sinuses when compared to the model where sinuses were not included.